Sub-Routine Posted January 6, 2008 Posted January 6, 2008 Now that I have the 99 and I am using a remote control I quickly decided that using the remote to control scenes is much better than controlling one device at a time. First I discovered that when a scene is on at it's default levels dimming and brightening maintain the same ratio between the lights in the scene until one or more devices reaches a limit, 0% or 100%. Then the ratios close until all devices are at the same level. If a scene is dimmed all the way off and brighten is called all the devices brighten at the same rate and there is no longer any differences that I would call a scene. What I did was create a scene with minimal brightnesses and call that scene on before I begin to brighten the scene. I put forth the proposition that as long as the ratios between the devices remain proportional to the initial scene that the scene is still true though brightened or dimmed. Rand
Mark Sanctuary Posted January 6, 2008 Posted January 6, 2008 Hey that’s a good idea Rand. I will add that to the definition above. Michel, if there is a tolerance it should be small because you want the check to be correct if something is not at its correct level or ratio.
Algorithm Posted January 6, 2008 Posted January 6, 2008 Each Controller has different scene levels so you would need to grab the levels for that node/controller in order to do these exact levels checks. I understood that from your original post, but still I am confused. If a scene has more than one controller, then isn't there more than one scene, i.e. one for each controller involved? The fact that they share the same devices means it is a common group, but still separate scenes, no? What I'm getting at is, wouldn't ISY maintain a separate scene status for each of those scenes?
Sub-Routine Posted January 6, 2008 Posted January 6, 2008 Michel, if there is a tolerance it should be small because you want the check to be correct if something is not at its correct level or ratio. When devices are instructed to brighten or dim they will use only 32 steps which means the resolution is 3.125%, not the 1% accuracy used in scene control.
Algorithm Posted January 6, 2008 Posted January 6, 2008 And, for the case of just one device on but, let's say, at +/-5% of the on level? I ask this simply because we cannot guarantee 100% accuracy unless we do a query. And, then, where do you draw the line? Perhaps I am misunderstanding Mark's intent; if so please set me straight, Mark. But I think the idea is that the status is based on ISY's predictive levels, i.e. what ISY thinks the levels are, rather than on doing an actual query. So, all the levels must exactly match ISY's predicated levels, and if those are within 5% (or whatever ISY's tolerance is) of actual levels, that's fine. Does that make any sense?
Sub-Routine Posted January 7, 2008 Posted January 7, 2008 Each Controller has different scene levels so you would need to grab the levels for that node/controller in order to do these exact levels checks. I understood that from your original post, but still I am confused. If a scene has more than one controller, then isn't there more than one scene, i.e. one for each controller involved? The fact that they share the same devices means it is a common group, but still separate scenes, no? What I'm getting at is, wouldn't ISY maintain a separate scene status for each of those scenes? Joe, You are right. Every controller has it's own scene. The way the ISY displays scenes is in imaginary groups. I like it. The ISY keeps a status of each responder no matter what scenes they are involved in. Rand
Mark Sanctuary Posted January 7, 2008 Posted January 7, 2008 If I do this is there a hidden tolerance? If Status 'Dining Hutch' is 10% Then what is the difference if it do this? If Status 'Dining Hutch' is 10% Or Status 'Family Desk' is 20% Or Status 'Family Main (load)' is 30% Or Status 'Family Screen' is not 40% Or Status 'Guest Bath Main' is 50% This is why I think the tolerance does not need to be big because it works already in the standard if statement. House Scene <-seperate levels 1 _____ Light _____ KPL (A) <-seperate levels 2 _____ KPL ( <-seperate levels 3 _____ Light _____ Light
Mark Sanctuary Posted January 7, 2008 Posted January 7, 2008 And also there would be six different ways to argument this command so some need to be precise and some need to be flexible.
Algorithm Posted January 7, 2008 Posted January 7, 2008 The ISY keeps a status of each responder no matter what scenes they are involved in. House Scene <-seperate levels 1_____ Light _____ KPL (A) <-seperate levels 2 _____ KPL ( <-seperate levels 3 _____ Light _____ Light Yes, you guys are right! Though I would consider them three different scenes, yet within ISY they are displayed within a single 'scene' (what I would call a 'group')--though it is actually the 'scene' for the ISY (Mark's 'separate levels 1'). But, I was always thinking that ISY would keep a status variable for each scene. It looks though that you are saying the status of any given scene would be evaluated each time it was requested--not 'queried', but simply evaluated based on the predictive level of each device of the scene. Is that right?
Mark Sanctuary Posted January 7, 2008 Posted January 7, 2008 This is how I think the code would go... [*:3gtmd3ul]Get all scene stored levels from selected node/controller[*:3gtmd3ul]Compare levels with the ISY 'live' levels (the table you see when selecting 'My Lighting')[*:3gtmd3ul]Repeat until all levels have been compared[*:3gtmd3ul]Return Yes/No Status
Mark Sanctuary Posted February 13, 2008 Posted February 13, 2008 Michel, Has this gotten any closer to the top of the to do list?
Michel Kohanim Posted February 14, 2008 Posted February 14, 2008 Hi Mark, No, 2.7 is at the top of our list. With kind regards, Michel Michel,Has this gotten any closer to the top of the to do list?
Mark Sanctuary Posted February 14, 2008 Posted February 14, 2008 Sounds good, just thought I would check.
Recommended Posts
Archived
This topic is now archived and is closed to further replies.